1

CONNECTIONS BETWEEN STRUCTURE AND PERFORMANCE OF FOUR 1

CATIONIC COPOLYMERS USED AS PHYSICALLY ADSORBED COATINGS 2

IN CAPILLARY ELECTROPHORESIS 3

4

Miguel Herrero1, José Bernal1, Diego Velasco2, Carlos Elvira2, 5

Alejandro Cifuentes1* 6

7

1Institute of Industrial Fermentations (CSIC), Juan de la Cierva 3, 28006 Madrid, Spain 8

2Institute of Science and Technology of Polymers (CSIC) Juan de la Cierva 3, 28006 9

Madrid, Spain 10

11

12

13

14

15

16

17

18

19

*corresponding author: 20

Tel# 34-915 618 806; Fax# 34-915 644 853 acifuentes@ifi.csic.es 21

22

2

ABSTRACT 23

In this work, the properties of four cationic copolymers synthesized in our laboratory 24

are studied as physically adsorbed coatings for capillary electrophoresis (CE). Namely, 25

the four copolymers investigated were poly(N-ethyl morpholine methacrylamide-co-26

N,N-dimethylacrylamide), poly(N-ethyl pyrrolidine methacrylate-co-N,N-27

dimethylacrylamide), poly(N-ethyl morpholine methacrylate-co-N,N-28

dimethylacrylamide) and poly(N-ethyl pyrrolidine methacrylamide-co-N,N-29

dimethylacrylamide). Capillaries were easily coated using these four different 30

macromolecules by simply flushing into the tubing an aqueous solution containing the 31

copolymer. The stability and reproducibility of each coating were tested for the same 32

day, different days and different capillaries. It is demonstrated that the use of these 33

coatings in CE can drastically reduce the analysis time, improve the resolution of the 34

separations or enhance the analysis repeatability at very acidic pH values compared to 35

bare silica columns. As an example, the analysis of an organic acids test mixture 36

revealed that the analysis time was reduced more than 6-times whereas the separation 37

efficiency was significantly increased to nearly 10-times attaining values up to 595000 38

plates/m using the coated capillaries. Moreover, it was shown that all the copolymers 39

used as coatings for CE allowed the separation of basic proteins by reducing their 40

adsorption onto the capillary wall. Links between their molecular structure, 41

physicochemical properties and their performance as coatings in CE are discussed. 42

43

Keywords: Coated capillaries/ physically adsorbed polymers/ proteins/ organic acids. 44

45

3

1. INTRODUCTION. 46

47

It is already well-known that capillary electrophoresis (CE) still shows some important 48

drawbacks that need to be overcome before CE can reach the large expectations that this 49

separation technique generated years ago. Thus, relatively low concentration sensitivity 50

with mostly used UV-absorption detection, low quantitative repeatability, migration 51

time drifts and the existence of analyte-capillary wall interactions are probably among 52

the most important problems still to be solved in CE. 53

54

In CE, the electrical charge on the capillary wall controls the zeta potential and the 55

electrostatic interactions, being responsible of electroosomotic flow (EOF) variations 56

[1] and interactions between solutes and capillary wall that ruin the CE separation, 57

mostly of compounds bearing a high positive charge [2-11]. 58

59

Although many different approaches have been proposed to solve these problems [12-60

17], so far, the procedure most commonly used to improve repeatability between 61

injections while reducing solute-capillary wall interactions is to coat the inner capillary 62

wall. Since the first CE coatings reported in 1985 [18], many different approaches have 63

been proposed [19] using covalently linked coatings [20-22] or physically adsorbed 64

coatings (static or dynamic) [23-34]. The development of new coatings for CE still 65

remains as an active area of research, as it is demonstrated through the publication of 66

several papers in 2010 [35-39], including several reviews on coatings used for CE-MS 67

[40] or for proteins separation [41]. 68

69

70

4

Our research group has been developing for years new coatings for CE [2-5,9,11,21,42-71

44], demonstrating its suitability for the CE analysis of different analytes including 72

peptides, proteins, DNAs, etc., or the compatibility of the coating with CE-MS [10]. As 73

many other laboratories usually do, we have mainly followed a trial-error approach in 74

order to obtain coatings with good CE capabilities. The novelty of the present work is 75

not only based on the synthesis and testing of two new copolymers presented for the 76

first time as CE coatings in this paper, but also on the comparative study among four 77

slightly different copolymers trying to extract possible correlations between their 78

physicochemical properties and their performance as CE coatings. Namely, in this 79

work, four different cationic copolymers synthesized at our laboratory are 80

comparatively investigated as physically adsorbed coatings in CE. The four copolymers 81

(whose structures and physicochemical properties are given in Figure 1 and Table 1) are 82

the following ones: MAEM/DMA, Poly(N ethyl morpholine methacrylamide–co–N,N-83

dimethyl-acrylamide); EPyM/DMA, Poly(N ethyl pyrrolidine methacrylate–co–N,N-84

dimethyl-acrylamide); EMM/DMA, Poly(N ethyl morpholine methacrylate–co–N,N-85

dimethyl-acrylamide); EPA/DMA, Poly(N ethyl pyrrolidine methacrylamide–co–N,N-86

dimethyl-acrylamide). As mentioned, EMM/DMA and EPA/DMA copolymers are 87

tested as CE coatings for the first time in this study. Advantages derived from the use of 88

the four coated capillaries including faster separation speed, better resolution, improved 89

repeatability and/or reduced analyte-capillary wall adsorption are discussed. 90

91

2. MATERIALS AND METHODS. 92

93

2.1. Samples and reagents. 94

95

5

Sodium hydroxide, acetone, formic acid and ammonium hydroxide were obtained from 96

Merck (Darmstadt, Germany). 3-[Cyclohexylamino]-1-propanesulfonic acid (CAPS) 97

and 2-[N-Morpholino]ethanesulfonic acid (MES), lysozyme (Mr ~14600; pI 10.9), 98

bovine cytochrome C (Mr ~12300; pI 10.5), horse cytochrome C (Mr ~12500; pI 10.2), 99

benzoic acid, L-ascorbic acid and sorbic acid, were supplied from Sigma-Aldrich 100

Chemie GmbH (Steinheim, Germany). Boric acid was purchased from Riedel-de Haën 101

(Hannover, Germany). Orthophosphoric acid was from Panreac (Barcelona, Spain). 102

Water was deionized with a Milli-Q system from Millipore (Bedford, MA, USA). 103

104

The monomers MAEM, EMM, EPA and EPyM were synthesized as described 105

elsewhere [45,46]. 2,2-Azobisisobutyronitrile (AIBN) from Fluka was purified by 106

fractional crystallization from ethanol. The solvents used were tetrahydrofuran (THF) 107

from Sharlau Chemie S.A. (Barcelona, Madrid) and ethanol from RP Normapur, 108

Prolabo (Fontenay, France). N,N-dimethylacrylamide (DMA) from Aldrich, was 109

vacuum distilled. Ammonium persulfate (Aldrich, Germany) was used as received. 110

Isopropanol (Scharlau, Scharlab S.L., Spain, 99.5% purity) diethyl ether (SDS, France, 111

99.7% purity) hexane (mixture of hexane isomers, SDS, France) tetrahydrofuran 112

(Scharlau, Spain, 99.8% purity) phosphorous pentoxide (Panreac, Spain, 98% purity), 113

N,N´-dimethyl formamide (DMF) (Scharlau, Scharlab, Spain, 99% purity). NaCl 114

(Panreac Química S.A., Spain, 99.5% purity) NaOH (Panreac Química S.A., Spain, 115

98%) and HCl (Analar Normapur, VWR International S.A.S., France, 37%) were used 116

for polymers synthesis and characterization. 117

118

The monomers MAEM, EMM, EPA and DMA were copolymerized by free radical 119

polymerization at 50º C using ammonium persulfate ([I] = 1.5 · 10-2 mol L-1) as radical 120

6

initiator and distilled water for MAEM [44] and a mixture of distilled 121

water/isopropanol (60/40) for EMM, EPA ([M] = 1 mol L-1) as solvent. After 24 hours 122

reaction time copolymer samples were dialyzed against water in a Spectra/Por® 123

(Spectrum Laboratories Inc., Houston, TX, USA) using membranes with a molecular 124

weight cut-off 3500, and then lyophilized. 125

126

The monomers EPyM and DMA were copolymerized [9] by free radical polymerization 127

at 50 ºC using AIBN ([I] = 1.5 · 10-2 mol L-1) as radical initiator and tetrahydrofuran 128

(THF) ([M] = 1 mol L-1) as solvent. The reaction mixture was precipitated in a large 129

excess of a diethyl ether-hexane mixture, purified by re-precipitation, filtered off and 130

vacuum-dried over phosphorous pentoxide. 131

132

133

2.2. Characterization of the copolymer systems. 134

135

The four copolymers were analyzed by 1H NMR spectroscopy using a Varian XLR-300 136

spectrometer operating at 300 MHz. All spectra were obtained at room temperature 137

from 5% (wt/v) CDCl3 and D2O solutions. 138

139

The pKa of the copolymers were determined by acid-base titration of 100 mg of 140

copolymer in 25 mL of a 0.1 M NaCl aqueous solution containing 2 mL of a 0.1 M HCl 141

solution to ensure the ionization of the amine groups of the copolymers. Diluted 142

aqueous solutions of NaOH were used to complete the titration using small volumes in 143

order to avoid the modification of the ionic strength. The changes of pH were measured 144

with a Schott GC841 pHmeter (Schott, Germany). Determination of the corresponding 145

7

compounds was carried out in triplicate. The standard deviation was in all the cases 146

lower than 3%. 147

148

Average relative molecular mass and relative molecular mass distributions were 149

determined by size exclusion chromatography (SEC) by using polymer solutions 150

(5mg/mL) in N,N - dimethyl formamide (DMF). Measurements were carried out at 1 151

mL/min flow using Ultrastyragel columns of 500, 104 and 105Å (Polymer Laboratories) 152

at 70 ºC and using a differential refractometer as detector. The calibration was 153

performed with poly (styrene) (PS) standards in the range of Mr ~ 2990 and Mr ~ 154

1,400,000 and polydispersity values lower than 1.1. 155

156

2.3. Preparation of background electrolytes (BGEs) for CE experiments. 157

158

To study the EOF mobility from the different coated capillaries at different pH, six 159

different background electrolytes (BGEs) were employed using either normal or 160

reversed polarity. These running buffer solutions were prepared by mixing adequate 161

volumes of the following solutions: 1 M H3PO4, 1 M HCOOH, 0.5 M MES, 0.25 M 162

H3BO3 and 0.25 M CAPS with 1 M NaOH. The final volume was adjusted to 50 ml, 163

resulting the next six running BGEs at the indicated pH values: (i) 86 mM 164

NaH2PO4/H3PO4 at pH 2.21; (ii) 75 mM NaHCOO/HCOOH at pH 3.60; (iii) 100 mM 165

Na-MES/MES at pH 6.19; (iv) 22 mM NaH2PO4/Na2HPO4 at pH 7.18; (v) 100 mM 166

NaB4O7/H3BO3 at pH 9.14; (vi) and 75 mM Na-CAPS/CAPS at pH 10.77. 167

168

2.4. Capillary coating procedure. 169

170

8

Four cationic copolymers composed by 80/20 EPyM/DMA, 80/20 MAEM/DMA, 80/20 171

EMM/DMA and 80/20 EPA/DMA were tested as capillary coatings for CE. Firstly, 1 172

mg of each copolymer was accurately weighted and dissolved in 5 ml of a solution 100 173

mM formic acid in water. Once the solution was perfectly dissolved, an equivalent 174

volume of 100 mM ammonium hydroxide was added in order to reach final copolymer 175

concentrations of 0.1 mg/ml. To get reproducible capillary coatings, a procedure already 176

developed in our group was applied [11]. Each new capillary was washed with 1 M 177

NaOH for 30 minutes and then flushed with the final copolymer solution for 15 min and 178

left to stand overnight at 4ºC. At the beginning of the day, the capillary was washed 179

with copolymer solution for 15 min in order to adequately prepare the coating. Between 180

runs, the capillary was washed using the next sequence: 3 min with copolymer solution, 181

2 min with water and finally 3 min using the selected running buffer. 182

183

2.5. CE separation conditions and samples. 184

185

A P/ACE 2050 instrument with UV detection from Beckman Coulter Inc. (Fullerton, 186

CA, USA) was used to perform the CE experiments. The capillaries used were made of 187

fused silica with 50 µm i.d. and 360 µm o.d. (Composite Metal Services, Worcester, 188

UK), with 370 mm of total length (300 mm detection length). For all the experiments 189

the capillary was thermostated at 25 ºC. 190

191

A solution of buffer/water/acetone (20/75/5) was employed as sample to measure the 192

EOF mobility; it was injected for 3 seconds at 3450 Pa and detected at 254 nm. The 193

voltage applied was 30 kV. Each run was performed by triplicate. For the study of the 194

9

different coatings, the same sequence of BGEs was used (i.e., from the lowest to the 195

highest pH) reproducing in that way any possible hysteresis effect. 196

197

Basic proteins sample consisted of lysozyme, bovine cytochrome C and horse 198

cytochrome C all dissolved in buffer (75 mM NaHCOO/HCOOH at pH 3.6)/water (1/4 199

v:v) at 75 µg/mL. The proteins were analyzed at 30 kV (normal polarity) with the bare 200

silica capillary and at -30 kV (reverse polarity) with the coated capillaries. The sample 201

was injected for 5 seconds at 3450 Pa; the detection wavelength was set at 214 nm. The 202

BGE employed was 75 mM NaHCOO/HCOOH at pH 3.6. 203

204

Organic acids test mixture contained benzoic acid, L-ascorbic acid and sorbic acid all 205

dissolved in buffer (75 mM NaHCOO/HCOOH at pH 3.6) /water (1/4 v:v) at 50 µg/mL. 206

The analysis of the test mix was performed at 20 kV (normal polarity) with the bare 207

silica capillary and at -20 kV (reverse polarity) with the coated capillaries. The detection 208

wavelength used was 280 nm, and the mixture was injected for 3 seconds at 3450 Pa. 209

The BGE used to analyze this test mix was 75 mM NaHCOO/HCOOH at pH 3.6. 210

211

3. RESULTS AND DISCUSSION. 212

213

3.1. Characterization of the copolymers 214

215

The characterization of the four copolymers was performed by 1H NMR spectroscopy 216

by the signal assignments shown in Figure 1. 217

218

10

Poly (MAEM-co-DMA): The MAEM and DMA feed molar fractions were 0.67 and 219

0.33 (corresponding to MAEM and DMA mass percentage of 80% and 20% 220

respectively). The signals used for the determination of composition of the copolymers 221

are: 0.83 (α-CH3, MAEM), 1.70 (-CH2- MAEM and DMA), 2.50 (3 × -CH2-N-, MAEM 222

cycle), 2.82 (2 × N-CH3 and –CH- of DMA), 3.12 (-CH2-NHCO- of MAEM) and 3.74 223

(-CH2-O-CH2- of MAEM cycle). The copolymers composition was determined by 224

integration of the corresponding signals being the MAEM molar fraction 0.69 and 0.31 225

for DMA. 226

227

Poly (EPyM-co-DMA): The EPyM and DMA feed molar fractions were 0.68 and 0.32 228

(corresponding to EPyM and DMA weight percentage of 80% and 20% respectively). 229

The signals used for the determination of composition of the copolymers are: 0.93 (α-230

CH3, EPyM), 1.60 (-CH2- EPyM and DMA), 1.70 (-CH2-CH2-, EPyM cycle), 2.50 (3 × 231

-CH2-N-, EPyM cycle), 2.82 (2 × N-CH3 and –CH- of DMA), 4.11 (-CH2-O-CO- of 232

EPyM). The copolymers composition was determined by integration of the 233

corresponding signals being the EPyM molar fraction 0.68 and 0.32 for DMA. 234

235

Poly (EMM-co-DMA): The EMM and DMA feed molar fractions were 0.67 and 0.33 236

(corresponding to EMM and DMA weight percentage of 80% and 20% respectively). 237

The signals used for the determination of composition of the copolymers are: 0.83 (α-238

CH3, EMM), 1.50 (-CH2- EMM and DMA), 2.50 (3 × -CH2-N-, EMM cycle), 2.82 (2 × 239

N-CH3 and –CH- of DMA), 3.74 (-CH2-O-CH2- of EMM cycle) and 4.11 (-CH2-O-CO- 240

of EMM). The copolymers composition was determined by integration of the 241

corresponding signals being the EMM molar fraction 0.73 and 0.27 for DMA. 242

243

11

Poly (EPA-co-DMA): The EPA and DMA feed molar fractions were 0.68 and 0.32 244

(corresponding to EPA and DMA weight percentage of 80% and 20% respectively). 245

The signals used for the determination of composition of the copolymers are: 0.93 (α-246

CH3, EPA), 1.60 (-CH2- EPA and DMA), 1.70 (-CH2-CH2-, EPA cycle), 2.50 (3 × -247

CH2-N-, EPA cycle), 2.82 (2 × N-CH3 and –CH- of DMA), 3.12 (-CH2-NHCO- of EPA. 248

The copolymers composition was determined by integration of the corresponding 249

signals being the EPA molar fraction 0.7 and 0.3 for DMA. 250

251

Table 1 shows some additional parameters of the synthesized copolymers whose 252

structures can be seen in Figure 1. Thus, as can be seen in Table 1, the reaction yields 253

(η) were in all cases higher than 80%. Monomer molar fractions in the feed (F) and in 254

the copolymer (f, determined by 1NMR spectroscopy) were found to be similar at high 255

conversions reactions. The ionisable character of the prepared polymers was studied by 256

the determination of their acid-base dissociation constant, pKa. The presence of tertiary 257

amine groups in all the positive monomers EMM, MAEM, EPA and EPyM (see Figure 258

1) confers the partial or total ionization of the polymers and with that their cationic 259

character (vide infra). Number average molecular weights determined by SEC were 260

found to be between 11000 and 34000 and polydispersity indexes determined by SEC 261

between 2.2 and 4.0, typical of radical polymerization. 262

263

264

3.2. Coated vs. bare fused silica capillaries. EOF and reproducibility studies. 265

266

In Figure 2, EOF is represented versus the pH of the background electrolyte (BGE) for 267

the bare silica capillary and for the four capillaries coated using copolymers containing 268

12

20% of the neutral monomer DMA and 80% of the positive monomer (MAEM, EPyM, 269

EMM and EPA). 270

271

According to the molecular structures of the four copolymers (see Figure 1), the amine 272

group in the monomers MAEM, EPyM, EMM and EPA is expected to provide some 273

cationic character to these macromolecules what can be of interest from a CE point of 274

view. Thus, as can be seen in Figure 2, the EOF obtained for the bare fused-silica shows 275

a typical dependence on the pH (i.e. EOF close to zero at very low pH, an increase of 276

EOF at pH about 5 and nearly a constant EOF value at pH values higher than 8). As can 277

be seen in Fig. 2, using the same running buffers with the coated capillaries the 278

behaviour of the EOF vs. pH is somewhat different. Thus, the coated capillaries show 279

an anodal EOF at low pH values, a nearly zero EOF at running buffer at pH 5-6 and a 280

low cathodal EOF at pHs higher than 8. This behaviour can be explained considering 281

that the global electrical charge onto the capillary wall is due to both the silanol groups 282

of the capillary (negatively charged) plus the amine groups of the polymer (bearing a 283

positive electrical charge). Thus, under acidic pHs the amine groups are the 284

predominant ionized groups bringing about a global positive charge and, as a 285

consequence, an anodal EOF. Under very basic pHs the number of positive charges on 286

the polymer decreases and the negative silanol groups become predominant, bringing 287

about a cathodal EOF. It is noteworthy that this EOF is reduced using the coated 288

capillaries what is a good indication of the shielding effect of the coating even at these 289

basic pHs. As can be deduced from Figure 2, the coating was successfully achieved in 290

all the cases as can be deduced from the negative EOF obtained at pH values lower than 291

5 using the four copolymers. 292

293

13

In order to study the reproducibility that can be achieved with these coatings, RSDs (% 294

RSD for n=5 injections) were calculated for the analysis time of the EOF obtained at six 295

different pHs from 2.21 to 10.77. Tables 2 and 3 show the RSD values obtained 296

depending on the pH and type of capillary used (four coated capillaries plus a bare silica 297

tube) for the same day, different days and different capillaries. Although the RSD 298

values were better using the bare fused-silica capillary at pHs above 6.19, it can be 299

observed that the use of coated capillaries clearly improved the RSD values at pH 2.21 300

and 3.60. This effect seems to be related to the magnitude of the EOF independently of 301

its direction towards the anode or the cathode (i.e., the higher the EOF value, the better 302

is the time reproducibility that can be achieved in CE). 303

304

305

3.3. Applications of coated capillaries. 306

307

In order to demonstrate the possibilities of using cationic copolymers as coatings in CE, 308

several applications were carried out. Thus, Figure 3 shows a comparison of the CE 309

separations of a group of organic acids achieved under identical conditions by using a 310

bare fused-silica capillary (Figure 3A) and four different coated capillaries (Figures 3B-311

E) at pH 3.60. As it can be seen, the use of the coated capillaries offers an interesting 312

advantage since they provide much faster separations at acidic pH values, making 313

possible the baseline separation of the three organic acids in less than 3 min compared 314

to the 18 min required by the bare silica tubing at the same pH. In this separation, due to 315

the EOF generated into the coated capillaries at the selected pH, the polarity was 316

reversed when using coated capillaries, what explains the different migration order 317

obtained. Besides, the separation was not only improved in terms of speed but also in 318

14

terms of separation efficiency probably due to the much lower molecular diffusion of 319

the organic acids in a fast separation. In fact, using the coated capillaries, efficiencies up 320

to 595000 plates/m were obtained for L-ascorbic acid. These values significantly 321

improved those obtained when using bare silica capillary that were below 64000 322

plates/m. Among the different copolymers employed as coatings, no great differences 323

were observed for this application. As it can be observed in Figure 2, both the 324

magnitude and the direction of the EOF were quite similar for all the coated capillaries 325

at the working pH (pH 3.60). 326

327

An additional advantage of these coated capillaries is the possibility to use them to carry 328

out separations of basic proteins in CE. One of the main problems during the separation 329

of basic proteins by CE is the electrostatic interaction between the positively charged 330

proteins and the negatively charged wall. Basic proteins can be separated at very low 331

pH, decreasing the degree of ionization of the silica surface. Alternatively, the capillary 332

can be coated to obtain a neutral or a positively charged surface allowing basic protein 333

separation. The use of cationic coatings as the ones presented in this work provides a 334

positively charged wall at pHs between 2 and 6 and whose EOF mobility can be tailored 335

to optimize resolution and separation speed by choosing the adequate polymer 336

composition and buffer pH. This is demonstrated in Figure 4, where three basic proteins 337

(lysozyme, horse cytochrome C, and bovine cytochrome C) are separated in a sodium 338

formate buffer at pH 3.6, using both bare silica capillary (Figure 4A) and the capillaries 339

coated using the four copolymers abovementioned (Figure 4B-E). As it can be observed 340

in Figure 4A, neither separation nor peaks were observed when using the bare silica 341

capillary due to the irreversible adsorption of these proteins onto the capillary wall. At 342

the same pH, the four coated capillaries provide high electroosmotic mobility (Fig. 2). 343

15

Therefore, these capillaries should provide at that pH a fast separation, but also a highly 344

positive charged surface, which should protect efficiently against proteins adsorption. 345

This fact is demonstrated in Figure 4, where the same three proteins at low 346

concentration (75 µg/mL) could be well separated in less than 6 min showing a 347

symmetrical peak shape in all the cases when coated capillaries were used. As it can be 348

observed in Figure 4, the migration behaviour was again similar using the four cationic 349

copolymers, although better resolution among the proteins were obtained using the 350

EMM/DMA coated capillary. Nevertheless, the fastest separation was obtained 351

employing the EPyM/DMA coated capillary which also provided good resolution 352

between the three proteins in less than 4.5 min. The efficiencies obtained for the four 353

coatings were slightly higher than 100000 plates/m. 354

355

356

3.4. Correlation between CE coating performance and copolymer structure 357

358

As it can be seen in Figure 1, the four copolymers of this work have in common the use 359

of the neutral monomer DMA varying the type of cationic monomer employed 360

(MAEM, EPyM, EMM and EPA). The four copolymers seem to behave similarly since 361

all of them are physically adsorbed onto the capillary wall and the coating is regenerated 362

between injections just by flushing the capillary with a dilute solution containing the 363

polymer. However, from the results given in Figures 2, 3 and 4, some interesting 364

differences among their behavior can be observed depending on their structure. In 365

general, the side chain group of the cationic monomer played an important role in the 366

CE behaviour of the coating, because copolymers will be ionised (or not) at a given pH 367

depending on the nature and pKa value of the mentioned group. This dissociation 368

16

constant value depends on the side chain group as can be deduced from Table 1 and 369

Figure 1 in which pyrrolidine (monomers EPA and EPyM) and morpholine (monomers 370

MAEM and EMM) based copolymers show pKa values close to 6 and 4, respectively. 371

Moreover, within the same type of side chain group the amide (used in MAEM and 372

EPA) or ester group (in EMM and EPyM) used to bind the side chain group to the main 373

polymer chain also can modify, although more slightly, the pKa value of these 374

macromolecules. This effect seems to be more intense in the case of pyrrolidine based 375

copolymers whose pKa values go from 5.7 using the amide group to 6.6 using the ester 376

group. As a result, pyrrolidine based copolymers show a stronger cationic character at 377

acidic pH values (as can be deduced from the higher cathodal EOF values obtained at 378

pH 2.21 as shown in Figure 2) and corroborated by the fast separations obtained using 379

EPA-DMA and EPyM-DMA copolymers (see Figures 3 and 4). Furthermore, the 380

higher pKa values of the pyrrolidine monomers brings about an important shielding 381

effect of the silanol groups at basic pHs, giving rise to lower EOF values as can be 382

deduced from the results of Figure 2. 383

384

385

4. CONCLUSIONS. 386

387

In this work, the possibility of attaining reproducible cationic coatings physically 388

adsorbed onto the capillary wall using MAEM/DMA, EPyM/DMA, EMM/DMA and 389

EPA/DMA copolymers was demonstrated. The two latter copolymers have been tested 390

for their use as coatings in CE for the first time in this study. The net charge of the 391

coating depends on the copolymer nature, copolymer composition and working pH, 392

determining the magnitude and direction of the EOF. Therefore, by modifying these 393

17

parameters the EOF can be effectively tailored to solve a given application in CE 394

introducing an additional and very useful parameter that can be optimized in CE. As it 395

has been shown in this work, the adequate selection of these conditions can: (i) improve 396

reproducibility; (ii) enhance both separation efficiency and resolution; (iii) increase 397

analysis speed and; (iv) diminish the negative adsorption between solutes and capillary 398

wall. 399

400

401

402

ACKNOWLEDGMENTS 403

404

M. H. and J. B. would like to thank Spanish Ministerio de Educación y Ciencia (MEC) 405

for their Juan de la Cierva contracts. This work was supported by Projects AGL2008-406

05108-C03-01 and CONSOLIDER INGENIO 2010 CSD2007-00063 FUN-C-FOOD 407

(MEC).408

18

REFERENCES. 409

[1] M. Chiari, M. Cretich, J. Horvath, Electrophoresis 21 (2000) 1521. 410

[2] F.B. Erim, A. Cifuentes, H. Poppe, J.C. Kraak, J. Chromatogr. A 708 (1995) 356. 411

[3] A. Cifuentes, J.C. Díez-Masa, J. Fritz, D. Anselmetti, A.E. Bruno, Anal. Chem. 70 412

(1998) 3458. 413

[4] A. Cifuentes, P. Canalejas, A. Ortega, J.C. Díez-Masa, J. Chromatogr. A 823 (1998) 414

561. 415

[5] A. Cifuentes, P. Canalejas, J.C. Díez-Masa, J. Chromatogr. A 830 (1999) 423. 416

[6] G.L. Erny, A. Cifuentes, Anal. Chem. 78 (2006) 7557. 417

[7] A. Carrasco-Pancorbo, A. Cifuentes, S. Cortacero-Ramírez, A. Segura-Carretero, A. 418

Fernández-Gutiérrez, Talanta 71 (2007) 397. 419

[8] G.L. Erny, M.L. Marina, A. Cifuentes, Electrophoresis 28 (2007) 4192. 420

[9] N. González, C. Elvira, J. San Román, A. Cifuentes, J. Chromatogr. A 1012 (2003) 421

95. 422

[10] C. Simó, C. Elvira, N. González, J. S. Román, C. Barbas, A. Cifuentes, 423

Electrophoresis 25 (2004) 2056. 424

[11] G.L. Erny, C. Elvira, J. San Román, A. Cifuentes, Electrophoresis 27 (2006) 1041. 425

[12] M.A. Hayes, I. Kheterpal, A. Ewing, Anal. Chem. 65 (1993) 27. 426

[13] J.S. Green, J.W. Jorgenson, J. Chromatogr. 478 (1989) 63. 427

[14] C. Schwer, E. Kenndler, Anal. Chem. 63 (1991) 1801. 428

[15] M.M. Bushey, J.W. Jorgenson, J. Chromatogr. 480 (1989) 301. 429

[16] N.A. Poldin, M.A. Hayes, Anal. Chem. 72 (2000) 1088. 430

[17] V. Kasicka, Z. Prusik, P. Sazelova, M. Chiari, I. Miksik, Z. Deyl, J. Chromatogr. B 431

741 (2000) 43. 432

[18] S. Hjerten, J. Chromatogr. 347 (1985) 191. 433

19

[19] C.A. Lucy, A. M. MacDonald, M.D. Gulcev, J. Chromatogr. A 1184 (2008) 81. 434

[20] Y.Wang, S. Hu, H. Li, N.L. Allbritton, C.E. Sims. J. Chromatogr. A 1004 (2003) 435

61. 436

[21] A. Cifuentes, J.M. Santos, M. de Frutos, J.C. Diez-Masa, J. Chromatogr. A 652 437

(1993) 161. 438

[22] R. Sebastiano, C. Gelfi, P.G. Righetti, A. Citterio. J. Chromatogr. A. 924 (2001) 439

71. 440

[23] M. Gilges, M.H. Kleemiss, G. Schomburg, Anal. Chem. 66 (1994) 2038. 441

[24] N. Iki, E.S. Yeung, J. Chromatogr. A 731 (1996) 273. 442

[25] C.L. Ng, H.K. Lee, S.F.Y. Li, J. Chromatogr. A 659 (1994) 427. 443

[26] M.N. Albarghouthi, T.M. Stein, A.E. Barron, Electrophoresis 24 (2003) 1166. 444

[27] M. Cretich, M. Stastna, A. Chrambach, M. Chiari, Electrophoresis 23 (2002) 2274. 445

[28] C.Q. Shou, C.L. Zhou, C.B. Zhao, Z.L. Zhang, G.B. Li, L.R. Chen, Talanta 63 446

(2004) 887. 447

[29] R.D. Sanzgiri, T.A. McKinnon, B.T. Cooper, Analyst 131 (2006) 1034. 448

[30] J.R. Catai, H.A. Tervahauta, G.J. de Jong, G.W. Somsen, J. Chromatogr. A 1083 449

(2005) 185. 450

[31] G. Danger, M. Ramonda, H. Cottet, Electrophoresis 28 (2007) 925. 451

[32] J. Znaleziona, J. Petr, R. Know, V. Maier, J. Sevcik, Chromatographia Suppl. 67 452

(2008) S5. 453

[33]S. Ullsten, A. Zuberovic, J. Bergquist, Methods Mol. Biol. 384 (2008) 631. 454

[34] J.K. Towns, F.E. Regnier, J. Chromatogr. 516 (1990) 69. 455

[35] J. Xu, L.Y. Yang, Z.F. Luo, Y.M. Wang, Electrophoresis 31 (2010) 1713. 456

[36] R.M. Yang, Y.H. Liu, C.Z. Zheng, J. Appl. Pol. Sci. 116 (2010) 3468. 457

20

[37] A. Elhamili, M. Wetterhall, M. Sjodin, R. Sebastiano, J. Bergquist, Electrophoresis 458

31 (2010) 1151. 459

[38] C.A. Lucy, I.M. Warner, Electrophoresis 31 (2010) 1036. 460

[39] Y. He, Y.M. Wei, X.H. Zheng, J.B. Zheng, Electrophoresis 31 (2010) 630. 461

[40] C. Huhn, R. Ramautar, M. Wuhrer, G.W. Somsen, Anal. Bioanal. Chem. 396 462

(2010) 297. 463

[41] Z. El Rassi, Electrophoresis 31 (2010) 174. 464

[42] A. Cifuentes, M. de Frutos, J.M. Santos, J.C. Díez-Masa. J. Chromatogr. A 655 465

(1993) 63. 466

[43] J. Bernal, I. Rodríguez-Meizoso, C. Elvira, E. Ibáñez, A. Cifuentes; J. Chromatogr. 467

A 1204 (2008) 104. 468

[44] J. Bernal, L. Sánchez-Hernández, C. Elvira, D. Velasco, E. Ibáñez, A. Cifuentes; J. 469

Sep. Sci. 32 (2009) 605. 470

[45] D. Velasco, C. Elvira, J. San Román, J. Mater. Sci. Mater. Med. 19 (2008) 1453. 471

[46] N. González, C. Elvira, J. San Román, J. Polym. Sci., Part A: Polym. Chem. 41 472

(2002) 395. 473

474

475

476

21

Table 1. Characterization of the copolymers used in this work: Sample (indicating 477

monomer mass percentage in the feed), F (molar fraction in the feed), f (molar fraction 478

in the copolymer), η (yield of the copolymerization reaction), pKa (acid-base 479

dissociation constant), Mn (number average molecular weight determined) and 480

polydispersity index determined by SEC. 481

482

Structure in Figure 1

Sample F f η (%) pKa Mn (Da)

Polydispersity index

A 80 MAEM

20 DMA

0.67 0.33

0.69

0.31 84% 4.2 11000 2.8

B 80 EPyM 20 DMA

0.68 0.32

0.68 0.32

82% 6.6 34000 2.2

C 80 EMM 20 DMA

0.67 0.33

0.73 0.27

82% 4.0 16000 3.7

D 80 EPA 20 DMA

0.68 0.32

0.70 0.30

83% 5.7 20000 4.0

483

22

Table 2. RSD (%) values of the EOF mobility for bare silica and different coated 484

capillaries at the different pH studied in this work. All copolymers used contained 20% 485

DMA. 486

487

488

489

490

491

492

493

494

495

496

497

498

499

pH RSD (%, n = 5) of EOF mobility values

Bare silica

MAEM /DMA EPyM/DMA EMM/DMA EPA/DMA

2.21 17.2 0.8 0.9 1.4 1.0 3.60 2.9 1.9 1.4 2.0 1.1 6.19 3.4 7.6 9.0 3.8 5.0 7.18 2.7 4.5 6.0 3.5 4.5 9.14 0.7 0.8 0.7 0.7 1.7 10.77 0.4 0.9 0.8 1.1 1.5

23

Table 3. Reproducibility (RSD, %) of the EOF mobility for three different days and three different capillaries using bare silica columns and 500

coated capillaries at the different pH studied in this work. All copolymers used contained 20% DMA. 501

502

pH Bare silica MAEM /DMA EPyM/DMA EPA/DMA EMM/DMA Three

consecutive days

Three different

capillaries

Three consecutive

days

Three different

capillaries

Three consecutive

days

Three different

capillaries

Three consecutive

days

Three different

capillaries

Three consecutive

days

Three different

capillaries 2.21 20.6 18.8 3.3 2.5 4.7 4.8 5.3 7.3 5.9 3.4 3.60 3.6 10.6 2.8 4.2 4.5 3.5 1.6 2.6 5.6 3.0 6.19 3.2 4.4 9.7 10.3 11.5 9.9 4.3 7.1 3.8 4.8 7.18 6.8 6.5 6.2 8.8 6.8 12.4 10.0 9.1 4.5 6.4 9.14 2.1 2.1 2.3 2.8 4.7 3.8 4.7 3.8 4.4 5.6 10.77 3.1 1.7 3.0 4.1 3.1 5.9 6.4 4.9 2.8 5.2 503

504

24

FIGURE CAPTIONS

Figure 1. 1H NMR spectroscopy of the four copolymers used in this work, namely: A:

MAEM/DMA, Poly (N ethyl morpholine methacrylamide–co–N,N-

dimethylacrylamide); B: EPyM/DMA, Poly (N ethyl pyrrolidine methacrylate–co–N,N-

dimethylacrylamide); C: EMM/ DMA, Poly (N ethyl morpholine methacrylate–co–

N,N-dimethylacrylamide); D: EPA/DMA, Poly (N ethyl pyrrolidine methacrylamide–

co–N,N-dimethylacrylamide).

Figure 2. Electroosmotic mobility as a function of pH and type of capillary.

Figure 3. Comparison of the separation of organic acids (1, benzoic acid; 2, L-ascorbic

acid; 3, sorbic acid) at 50 µg/ml in A) bare silica capillary, B) 80:20 MAEM/DMA

coated capillary, C) 80:20 EPyM/DMA coated capillary, D) 80:20 EPA/DMA coated

capillary, and E) 80:20 EMM/DMA coated capillary. Separation conditions: capillary

length, 37 cm; effective length, 30 cm; capillary i.d., 50 µm; background electrolyte, 75

mM NaHCOO/HCOOH at pH 3.60; voltage, -20 kV (except A, +20 kV).

Figure 4. Comparison of the separation of basic proteins (1, lysozyme; 2, bovine

cytochrome C; 3, horse cytochrome C) at 75 µg/ml in A) bare silica capillary, B) 80:20

MAEM/DMA coated capillary, C) 80:20 EPyM/DMA coated capillary, D) 80:20

EPA/DMA coated capillary, E) 80:20 EMM/DMA coated capilary. Separation

conditions: capillary length, 37 cm; effective length, 30 cm; capillary i.d., 50 µm;

background electrolyte, 75 mM NaHCOO/HCOOH at pH 3.60; voltage, -30 kV (except

A, +30 kV).